Tuesday, August 25, 2015

Snow is white

I frequently see the idea of meaning explained like this:

The meaning of "snow is white" is that snow is white.

Authors who write that should be ashamed of themselves. 

For one thing they are trying to define meaning as the removal of quotation marks. That does not define any meaning when there are no quotation marks and could lead to the impression that only quoted strings can have a meaning; which is absurd. (Adding further quotation marks doesn't help but you can make some progress if you use a notation for the inverse of quotation marks.)

For another: Just how lazy are you? A more correct answer is that, there is a thing called "snow" with an attribute called "white", so the structure of the meaning is shown as
snow_/white
The content of the meaning consists of  the many contextual specifics associated to "snow" and "white". For example that white is a color or that snow is cold and happens in winter. If you want to try annotating it, there is a huge web of implicit meaning surrounding the expression.

Sunday, August 9, 2015

Prosody and Truism

I  assume that different language mechanisms - gesture, prosody, word and expression meanings, all work together in parallel. So for a lot of simple standard expressions, with different meanings, there should be different mechanisms of gesture and prosody that go along with them.

This thought had me wondering if something as simple as a truism would not correspond with something simple in prosody. After all, truisms embody very standard implicit thoughts. Why shouldn't they have tunes to go with them?

So I found one and think it important. Truism 7 says: "what is contrasted with is later allowed" or X*::X. This has the musical counterpart of "breath in - breath out" or G7-C. The first creates tension and the second relieves it. It is important to note that James's version of Hegel has the truism running in either directions as a "double negative". I don't know about Hegel, but think James is incorrect that the formula is bi-directional. It has to start with tension and end with relief. The other way around doesn't work in language or in music, as it leaves one in suspense.

This is an example where a semantic theory could contribute to a philosophical discussion. The main contribution would be in filtering out the non-philosophic "semantics" from the discussion.

Reading William James: Hegel and his Method

I am reading James and he relates Hegel's weirdly phrased notion that: an object completes itself with its own negation. James goes on with several pages of discussion leading to a list of examples:
 "Extreme anarchy and extreme despotism lead to one another. Pride comes before a fall. Too much wit outwits itself. Joy brings tears, melancholy a sardonic smile."

But actually Hegel's notion (or at least Jame's version of Hegel which I do not think is quite right) is not so strange. Those phrases are all versions of Truism 7: X*::X (or flouting this truism).

The neat thing about semantics is that it can be a very concrete empirical science that studies real world words, sentences, paragraphs, etc. To assert that a rule like Truism 7 is assumed implicitly is a concrete observation, easy to demonstrate with simple examples (like James's list).  Its validity as a truism can be analyzed or disputed much more concretely than how one would analyze or dispute Hegel. At the same time, expressing such ideas as language phenomena frees up philosophers either to do other work or else focus more on the connection of their problems to language.

James also complains that Hegel's vagueness leaves his reader always confused as to whether Hegel is writing about physics, psychology, or philosophy. It makes broad sense if Hegel is talking about the mind that studies those subjects. To me it is gratifying to simply say those subjects are semantically constrained, just like Newton Laws of Motion..

For the record I do not think Truism 7 is the equivalent of Hegel's dialectic. My understanding of the latter is that it had to do with the act of becoming meta (going up one level of abstraction) and how when that act turns upon itself and is reabsorbed back into the original perception, is when and how we learn about the object.

Saturday, August 8, 2015

Does anyone read this blog?

Google Analytics seems to think someone visits this blog. But I never saw a comment from anyone. I would be shocked to know someone was actually reading this stuff about semantics. I would be even more shocked if someone out there wanted to communicate about it. So: is there anybody out there?
Update: Until I see a comment from someone, I'll assume readers do not exist.
Update: In fact, the first week after I went on Google Analytics (goodbye Sitemeter, you were good to me and more useful than Google Analytics) there was a bulge of 50 or so daily visits. But I think those were just bots.

Theorizing about Semantics (it's a waste of time)

I have to imagine that contributions to semantics are themselves a form of language use that is about as abstract as possible. So I am poking around on the internet and wanting to read contributions to semantics and instead I find authors, one after the other, discussing the validity of one form of semantics versus another. No example of language is analyzed; no new approaches are revealed. They found something even more abstract to talk about!

I have long thought the translation of expressions into "propositions" and the analysis of their "truth properties" was a distraction from the qualification of the meaning of an expression. Usually, expressions mean something about the world - in its wonderful diversity. Little is added in saying the meaning is "true", because this auxiliary meta attribute is not a substitute for the thing itself. Also, very few expressions can be considered "true" - for example anything in a novel.

Rather than getting more and more abstract - going from worldly meaning to "truth", to the meta discussion of semantics, to a meta-meta discussion of what is the "very best-est" of semantic theories  - I regard these as digressions taking us further and further from the subject. I want to study the articulation of the world as expressed by people. You cannot do that if you spend your time evaluating the semantic theories of other authors.

Though, ironically, I want to ignore my own advice and say: lots of authors are looking at the same set of topics as I am but organizing their ideas in a way I find awkward. It is not that they are wrong but that they are clumsy. But I want to say I think modern semantics is mostly off in the weeds of "meta discussion", doing the wrong kind of work and not advancing us towards a goal of automated language comprehension.

Wednesday, July 29, 2015

Vagueness in the Truism Classification

It is a bit confusing to have Truism 6N cover the act of making implicit assumptions into explicit parts of a narrative, while Truism 1 and 5N involve implicit assumptions. Implicit assumptions can either (a) be used in analysis of a sentence; or (b) become part of the explicit narrative.

Truism 1 states implicit assumptions about events.
Truism 5N says that facts are implicit assumptions.
Truism 6N says any implicit assumptions (not just 1 and 5N) can be made explicit.

I remain confused about this because I am not sure what are all the implications for an implicit assumption. It generates an optional narrative fragment that may appear explicitly (6N) or it can remain implicit and only appear in meta discussion.

Or, a truism may make sense of a later part of a sentence, or help anticipate it. Two different uses?

One more clarification (you can tell I am unsure): Truism 6N is N([Z],Z where the connector is a ',' and not a '::'. This is important because there is no automatic use of "but" to reverse polarities. So Trusim 1 is not expected to have a "but"-dependent polarity. And Truism 5N makes no assumptions about what known factual narratives it is referring to. I like the word "flout" which they use to indicate polarity reversal for "implicature" - an idea close to that of "truism" and very much the same as the affect of "but".

Monday, July 27, 2015

Truism 7: What has been contrasted with is now allowed

This takes a very general form.
Truism 7: What has been contrasted with is now allowed.
X* :: X
I had a lot of trouble deciding (a) that this is a truism that is used; (b) that it is used even in the language of the concrete events and persons.

Rationale: At first I noticed this truism being used in literature. For example the Clive Cussler sentence (see here) begins with something like "He almost fell asleep". I noted that, by the end of the paragraph, he does finally get to go to sleep. So the whole paragraph might be summarized by this truism:
(he_/sleep)* :: (he_/sleep)
More recently I noticed the truism "you get what you want" was mis-classified in terms of a "want" verb. I decided that "want" describes an experience of lacking something, hence it is related to the adjectives of "having"/"not-having" and those, in turn, are adjectives modified by the verb pair "get"/"lose". Making those verbs more primary and putting them into my "starter" verb table, meant taking out "want". In the new analysis "want" is contrasted with having. So "X wants Y" is translated as: 
(X_/has Y)*
In that case the "you get what you want" is a simple application of Truism 7 to this particular adjective narrative:
(X_/has Y)* :: (X_/has Y)
Example: John was hungry so he went out to eat."